Some tasks are inferred based on the benchmarks list.
The benchmarks section lists all benchmarks using a given dataset or any of its variants. We use variants to distinguish between results evaluated on slightly different versions of the same dataset. For example, ImageNet 32⨉32 and ImageNet 64⨉64 are variants of the ImageNet dataset.
In order to fill the gap of HC3 under semanticinvariant tasks, we extend HC3 and propose a larger ChatGPT-generated text dataset covering translation, summarization, and paraphrasing tasks, called HC3 Plus.